A DV AN CE
ARBEITSPHYSIOLOGISCH-BIOMECHANISCHE ANALYSE EINES PASSIVEN
EXOSKELETTS ZUR UNTERSTÜTZUNG VON BERUFLICHEN HEBE- UND
BEUGEVORGÄNGEN
A DV A NC E
WORK PHYSIOLOGICAL-BIOMECHANICAL ANALYSIS OF A PASSIVE
EXOSKELETON TO SUPPORT OCCUPATIONAL LIFTING AND FLEXING
PROCESSES
FOR REVIEW BY THE
ETHICS COMMITTEE
OF THE
EBERHARD-KARLS-UNIVERSITY OF TÜBINGEN
Date of protocol 17.01.2019
Version of protocol V3
Head of study (UKT) Dr. Benjamin Steinhilber & Dr. Tessy Luger &
Dipl.-Ing. Robert Seibt, Mona Bär M.Sc.
Institute of Occupational and Social Medicine and Health
Services Research
Wilhelmstraße 27
72074 Tübingen
℡ 07071-29-86805 | � 07071-29-4362
Director Prof. Dr. Monika A. Rieger
Other researchers Sylvia Weymann (Bachelor Medicine Technique)
Gianluca Caputo (Bachelor Medicine Technique)
Research Proposal ADVANCE Page i
I Abstract
Introduction: Industrial tasks that are characterized by high loads, a high repetition rate, and / or
awkward body postures, put employees at higher risk to develop work-related musculoskeletal
disorders (WRMSD), especially low back pain. To counteract the prevalence of WRMSD, human-robot
interaction could improve the power of a person and reduce the physical strain. For the lower back, a
reduction of spinal loading could be helpful. The passive upper-extremity exoskeleton Laevo® is
developed to support physically heavy work: it supports the back during bending and should,
consequently, result in less low back pain (Laevo®, the Netherlands).
Objectives: The primary aim of this study is to assess to what extent wearing the exoskeleton
changes knee compression force, reduces muscular load on the back muscles, and changes the
muscular load on other body parts such as the legs and shoulders in different tasks (static vs.
dynamic) and different postures (trunk flexion vs. trunk flexion and rotation). The secondary aim of
this study is to investigate the relation between trunk flexion angle, load carried and muscular stress
of the back muscle. The tertiary aim of this study is to estimate internal loadings on the spine, using a
lumbar spine model. The quaternary aim of this study is to assess the performance of subjects during
functional activities (e.g., stair climbing) when wearing the exoskeleton.
Methods: For the primary aim, we will test six different experimental conditions in the laboratory,
which are a combination of exoskeleton (with vs. without Laevo®), task (static vs. dynamic), and knee
angle only for the dynamic task (flexed vs. extended). Within each combination, we will test three
different working directions (front vs. left vs. right), realized by changing the working posture (trunk
flexion vs. left trunk rotation vs. right trunk rotation). Using the single Williams design for six
conditions, we estimated our sample size to include 36 subjects (i.e., a multiple of six). Using a force
plate, acceleration and postural sensors, we can estimate knee compression force using 2D inverse
modelling. With an electromyographic system, we can record the muscle activity of selected target
muscles at different body parts (i.e., legs, trunk, and shoulders). The heart rate will be recorded using
electrocardiography. For the secondary aim, we will test four different conditions, which are a
combination of exoskeleton (with vs. without Laevo®) and knee angle (flexed vs. extended). Within
each combination, we will test three different loads carried (0kg, 8kg, 16kg) and five different trunk
flexion angles (0°, 30°, 60°, 60°, 30°). We will record muscle activity, position, heart rate and ground
reaction forces. For the tertiary aim, we will use the lumbar spine model developed by the research
group Biomechanics and Biorobotics of the research cluster Simulation Technology of the University
of Stuttgart. The model includes a detailed lumber spine with non-linear discs, ligaments, and
muscles. Using the measurements of the experiment, this model is able to predict how internal
forces in the lumbar spine change as a result of external forces (i.e., wearing and using the Laevo®
exoskeleton). For the quaternary aim, we will test three different functional tests. The main outcome
for the quaternary aim is performance, which could be either time recorded or difficulty rated on a
standardized 11-point numeric rating scale.
Analyses: For the primary aim, the effects of exoskeleton (with vs. without), task (static vs. dynamic),
knee angle (flexed vs. extended; only for the dynamic task), and working posture (trunk flexion vs.
left trunk rotation vs. right trunk rotation) will be assessed using a four-factor repeated-measures
analysis of variance (RM-ANOVA) or a generalized estimating equation (GEE) which is more robust.
Research Proposal ADVANCE Page ii
For the secondary aim, the effects of exoskeleton (with vs. without), knee angle (flexed vs.
extended), load carrier (0kg vs. 8kg vs. 16kg), and trunk flexion angle (0° vs. 30° vs. 60°) will be
assessed using a RM-ANOVA or GEE. For the quaternary aim, the effect of exoskeleton (with vs.
without) on performance of each functional test will be assessed using a paired T-Test.
Data protection: All participating subjects will receive a refund of € 45 after study completion.
Subjects will sign an informed consent and their data will be numerically pseudonymized to
guarantee anonymity.
II Keywords
Passive exoskeleton; forward trunk flexion; trunk rotation; inverse dynamics; muscle activity;
modelling
Research Proposal ADVANCE Page iii
III Attachments
Attachment 1: Experimentelle Bedingungen, randomisiert
Attachment 2: Anwerbetext zur Studie
Attachment 3: Probandeninformation
Attachment 4A: Einwilligungserklärung
Attachment 4B: Datenschutzerklärung
Attachment 5: Ein- und Ausschlusskriterien
Attachment 6: Allgemeine Angaben
Attachment 7: Nordische Fragebogen
Attachment 8: Selbsteinschätzung von Beschwerden
Attachment 9: Beschwerden Intensität & Körperregion
Attachment 10: NASA-TLX
Attachment 11: Bewertungsfragebogen
Attachment 12: EU-Konformitätserklärung – Laevo® Exoskelett | Laevo B.V.
Attachment 13: EU-Konformitätserklärung – PS11 Messgeräte | Thumedi GmbH & Co. KG
Attachment 14: EU-Konformitätserklärung – PS12 Messgeräte | Thumedi GmbH & Co. KG
Attachment 15: EU-Konformitätserklärung – Force Plate | Bertec GmbH
Attachment 16: EU-Konformitätserklärung – Force Sensor | ME Meßsysteme GmbH
Research Proposal ADVANCE Page iv
Table of content
I AB STRACT I
I I KEYWORDS II
I I I ATTACHM ENTS III
1 ABBREVIATIONS 1
2 BACKGROUND 2
3 OB JECTIVES 4
4 STUDY DESIGN AND STUDY POPULATION 5
4.1 ST U D Y D E S I G N 5
4.1.1 Independent variables 5
4.1.2 Outco mes 5
4.2 ST U D Y P O P U L A T I O N 5
4.3 IN - A N D E X C L U S I O N C R I T E R I A 6
4.3.1 Inc lusio n c r i ter ia 6
4.3.2 Exc lusio n c r i ter ia 6
4.4 RE C R U I T M E N T 7
5 TIM ETABLE 8
5.1 WP 1: S T A T I C W O R K T A S K S 8
5.2 WP 2: D Y N A M I C W O R K T A S K S 9
5.3 WP 3: L U M B A R S P I N E M O D E L 9
6 STUDY PROCEDURE 10
6.1 E X P E R I M E N T A L T A S K S 10
6.2 MO D E L L I N G O F T H E L U M B A R S P I N E 11
6.3 ST A T I C T R U N K F O R W A R D F L E X I O N A N G L E W H I L E H O L D I N G A L O A D 12
6.4 FU N C T I O N A L T A S K S 13
Research Proposal ADVANCE Page v
6.5 E X P E R I M E N T A L P R O C E D U R E 13
6.6 OU T P U T 14
7 MEASUREM ENT M ETHODS AND ANALYSES 15
7.1 GE N E R A L D A T A 15
7.2 MU S C L E A C T I V I T Y 15
7.3 K I N E M A T I C S : P O S T U R E A N D F O R C E 15
7.4 RE F E R E N C E M E A S U R E M E N T S 17
7.5 HE A R T R A T E 17
7.6 D I S C O M F O R T 17
7.7 FU N C T I O N A L T E S T S 18
7.8 RE L A T I O N S T A T I C T R U N K F O R W A R D F L E X I O N A N G L E , C A R R I E D L O A D A N D M U S C U L A R S T R E S S /
S P I N A L L O A D 18
7.9 PA R T I C I P A N T E V A L U A T I O N 18
7.11 ST A T I S T I C A L A N A L Y S I S 19
8 RISK FOR THE PARTIC IPANT 20
9 DATA PROTECTION 21
10 STUDY F INANCE 22
11 PERSONNEL STAFF 23
11.1 WO R K P A C K A G E S 1 & 2 23
11.2 WO R K P A C K A G E 3 23
11 REFERENCES 24
Research Proposal ADVANCE Page 1
1 Abbreviations
%RVE percentage of the reference voluntary electrical activity
2D two-dimensional
3D three-dimensional
ANOVA analysis of variance
EMG electromyography
�⃗ force
�⃗ gravitational force
IED inter-electrode distance
L5 fifth lumbar vertebra
LCT Lift and Carry Test
M moment
MPF median power frequency
� angle
RPD rating of perceived discomfort
RMS root-mean-square
RVC reference voluntary contraction
RVE reference voluntary electrical activity
SCT Stair Climbing Test
SimTech Simulation Technology
SLS Single-Leg Stance
SUS System Usability Scale
T3 third thoracic vertebra
TAM2 Technology Acceptance Model 2
UKT University Hospital Tübingen
WP work package
WRMSD work-related musculoskeletal disorders
Research Proposal ADVANCE Page 2
2 Background
Due to the exposure to physically demanding work, repetitive movements and awkward body
postures in industry (van Rijn et al., 2010), a considerable amount of employees develops work-
related musculoskeletal disorders (WRMSD; Wang et al., 2017). Low back pain is one of the most
prevalent disorders in the general population (Scott et al., 2010), but is also apparent as WRMSD
(Irwin et al., 2007). Because industry cannot be completely automated, human-robot cooperation is a
solution to provide assistance or support for human motions or actions, but does not influence the
flexibility or creativity of motions and actions (MacDougall, 2014).
One type of human-robot cooperation is the exoskeleton, which is a wearable, external system
improving the power of a person (de Looze et al., 2015). Exoskeletons can be distinguished by their
technical functions, i.e. as an active or a passive system and by its target body part it supports.
Exoskeletons are mainly used for rehabilitation or other medical purposes supporting daily activities
like walking, standing up, or grasping (Viteckova et al., 2013). However, exoskeletons in occupational
settings are not often used, mainly because there are no specific guidelines available for safety or
technical issues. This means that exoskeletons will not yet be implemented on a large scale in
industry to assist work-related activities. Aspects such as perceived discomfort, body alignment and
weight of the exoskeleton are important aspects that need to be considered when optimizing
exoskeletons for implementation in occupational settings. However, several reviews have been
conducted already, including a review on the two key applications of exoskeleton systems (Yang et
al., 2008), on the characteristics of exoskeleton robot systems for power assistance (Lee et al., 2012),
on the compatibility of lower limb exoskeletons with the user (Viteckova et al., 2013), and on the
effectiveness of wearing exoskeletons on the physical loading of the body (de Looze et al., 2015).
The review of Yang et al. (2008) gives a great overview of the main two fields of application of
exoskeletons, which are teleoperation and power amplification. Lee et al. (2012) in their review
report on the different classifications of exoskeletons, including (1) power assistant or power
augmented systems, (2) anthropomorphic, quasi-anthropomorphic, or non-anthropomorphic, (3)
active, passive, or quasi-passive, and (4) the level of human-robot interactions. Both reviews point
out that a lot still needs to be achieved with regard to the level of the exoskeleton’s
anthropomorphicity, as it appears to be very difficult to mechanically pursue natural human motion.
Viteckova et al. (2013) add to this challenge the problem of the high energy consumption of active
exoskeletons and the composition of the exoskeletons which, with a high weight, require high
metabolic consumption of the wearer. The review of de Looze et al. (2015) was the only one
specifically evaluating the effect of wearing exoskeletons on the physical loading of the wearer’s
body. They found that muscular activity was reduced by 10-40% when wearing passive systems and
up to 80% when wearing active systems. However, they also point out that several challenges need
to be overcome, including degeneration of the exoskeleton’s target muscles, high levels of
discomfort and increased physical load on other than the exoskeleton’s target body parts, and true
impact on potentially reducing injury prevalence (de Looze et al., 2015).
Overall, exoskeletons need to be tested in highly controlled situations, because many practitioners
see exoskeletons as a chance to optimize ergonomic work design. One way to test exoskeletons is in
laboratory studies, evaluating their direct influence on the wearer and the work performance.
Recently, the passive upper-body exoskeleton Laevo® has been studied in detail (Laevo®, Delft, the
Research Proposal ADVANCE Page 3
Netherlands). Bosch et al. (2016) assessed the effect of wearing the Leavo® exoskeleton on the
activity of the back muscles during trunk forward flexion and possible negative side effects for other
muscles such as the abdominal and hip extensor muscles. They simulated an assembly task (10 work
cycles of 20-30 s) and a static posture (endurance time up to 9 min) while the trunk was in 40°
forward flexion. The tasks were performed with and without wearing the Laevo® exoskeleton. As
expected, the lower back muscle activity decreased in both tasks with 35-44%. Additional
measurements showed that the muscle activity decreased in the upper back with 44-50% and in the
hip extensor with 20-24%. The activity of the abdomen muscles did not change when wearing the
Laevo®. Despite these considerable decreases in muscular activity, the perceived discomfort in the
chest increased, because this is the place where the exoskeleton exerts pressure to the body. Bosch
et al. (2016) also observed that users wearing the Laevo® exoskeleton tend to overstretch their
knees. This overstretch may be due to the pressure of the chest pads on the body, which pushes the
upper body backward and hampers knee flexion.
The study of Bosch et al. (2016) is a good first step to map the operation of the system and its effects
on the physical load of the user. The results are promising with respect to muscular activity at the
upper body, but unclear is what happens to the compression forces of the knee, what happens to the
user’s posture during dynamic work tasks, and what happens to the physical load when the user is
diverting from the symmetry line (sagittal plane) while bending. This information is necessary to
assess the influence of this exoskeleton on the body, prior to a field feasibility study and its long-term
implementation and use in the field.
Research Proposal ADVANCE Page 4
3 Objectives
This study has three objectives:
1. Assess to what extent wearing the Laevo® exoskeleton (with vs. without) changes knee
compression force, reduces muscular load on the back muscles, changes muscular load on
other body parts such as the legs and shoulders, and changes discomfort in different tasks
(static vs. dynamic) and different postures (trunk flexion vs. trunk flexion and rotation).
2. Estimate internal loadings on the spine, using a lumbar spine model, and see what the
influence is of wearing the Laevo® exoskeleton.
3. Assess the performance of subjects during functional activities (e.g., stair climbing) when
wearing the Laevo® exoskeleton (with vs. without).
Research Proposal ADVANCE Page 5
4 Study design and study population
4.1 Study design
4.1.1 Independent variables
In the study, three independent variables will be investigated:
- Exoskeleton: with vs. without;
- Task: static vs. dynamic;
- Knee angle: flexed vs. extended (only for the dynamic task)
- Working posture: straight lifting/bending vs. lifting/bending with left/right trunk rotation.
4.1.2 Outcomes
In each experimental condition, the following
outcomes will be assessed:
- Muscular loading;
- Knee joint loading;
- Force on legs and chest due to the
exoskeleton’s pads;
- Performance on functional tests;
- Subjective rating of perceived
discomfort;
- Participant evaluation of the
exoskeleton;
- Spinal loading.
4.2 Study population
We will select healthy, male participants in
the age of 18-40 years old. For the main
experiment, we constructed a single Williams
design (Bate and Jones, 2006) based on three
independent variables with two levels each
that resulted in six different orders of
conditions (Figure 1A). The single Williams
design corrects for possible first-order carry-
over effects. Using the results published by
Bosch et al. (2016), we estimated the sample
size including only the comparison of with
and without Laevo® exoskeleton. Based on
the reported outcomes of back discomfort,
we estimate a sample size of 29 subjects.
Combining both calculations, we strive to
include a multiply of the single Williams
design, i.e. a multiple of six, which would be
30. However, we prefer to include 6x6
Figure 1A. The single 6×6 Williams design, based on a
balanced, uniform cross-over design.
Figure 1B. The single 4×4 Williams design, based on a
balanced, uniform cross-over design.
Figure 1C. The double 3×3 Williams design, based on a
balanced, uniform cross-over design.
1 2 3 4 5 6
1 A B F C E D
2 B C A D F E
3 C D B E A F
4 D E C F B A
5 E F D A C B
6 F A E B D C
Subject
Period
1 2 3 4
1 A B D C
2 B C A D
3 C D B A
4 D A C B
Subject
Period
1 2 3
1 A B C
2 B C A
3 C A B
4 C B A
5 A C B
6 B A C
Period
Subject
Research Proposal ADVANCE Page 6
subjects, leading to a number of 36 subjects (Attachment 1). The order of the three working
positions (forward trunk flexion vs. left trunk rotation vs. right trunk rotation) will be randomized
separately, using a double Williams design, i.e. 6 different orders.
For the additional postural angle and weight measurements that will be performed in a full-factorial
design, we will measurements consisting of exoskeleton status (with or without Laevo®), knee angle
(flexed, extended), static trunk forward flexion angle (0°, 30°, 60°, 60°, 30°), and carried weight (0kg,
8kg, 16kg). The conditions will be offered in a randomized order, based on the two independent
variables exoskeleton status and knee angle, generated by a single Williams design with four
different orders (Figure 1B). The static trunk flexion angles are always performed in ascending and
descending order, i.e. 0°, then 30°, then 60°, then 60°, then 30°, which does not need randomization.
The order of the three carried weights will be randomized separately, like the working positions, i.e.
by using a double Williams design leading to 6 different orders (Figure 1C).
For the additional functional tests, which are three (i.e., Stair Climb Test [SCT], Timed Up-and-go Test
[TUG], course) and always performed in the same order, we randomized by exoskeleton status by
drawing lots.
4.3 In- and exclusion criteria
4.3.1 Inclusion criteria
- Age between 18 and 40 years;
- Male;
- BMI < 30 kg/m2;
- The participant will give his voluntary informed consent after receiving oral and written
information of the content and goal of the study.
4.3.2 Exclusion criteria
- Aged <18 and >40 years;
- Female;
- People under the influence of intoxicants, analgesics, or muscle relaxants;
- Alcohol abuse;
- People with cardiovascular diseases;
- People with a heart pacemaker;
- People with a disability who, due to their restriction at a workplace of this kind, will not be
able to participate;
- People with Diabetes Mellitus;
- People with severe muscle contractions of the lower extremities, back or arms;
- People with acute ailments or pain;
- People who are unable to complete the examination program due to language or cognitive
obstacles;
- Depending on the degree of severity, people with diseases of the veins and joints of the
lower extremities, spine, muscle disorders, symptomatic neurological-psychiatric diseases,
acute pain syndromes, maladies or other current diseases.
Research Proposal ADVANCE Page 7
4.4 Recruitment
Subject will be recruited by means of asking fellow colleagues / students in Tübingen to voluntary
study participation. When necessary, additional announcement e-mails will be sent to employees /
students from the University and Hospital of Tübingen (UKT), and to former subjects who agreed to
be informed about future studies (Attachment 2), emphasizing the voluntary principle.
Research Proposal ADVANCE Page 8
5 Timetable
The project has three overarching work packages (WP; Table 1), each containing a number of smaller
work packages. The complete duration of the project is 14 months, with study start in July 2018,
which includes preparations for the Ethical Committee.
Table 1. The timetable in months for the proposed study that is intended to start in July 2018. Numbers represent the work
package (WP) and letters represent the smaller sub-work package.
Year 2018 2019
Month
Jul
Au
g
Se
p
Oct
No
v
De
c
Jan
Fe
b
Ma
r
Ap
r
Ma
y
Jun
Jul
Au
g
1
1A
1B
1C
1D
1E
1F
1G
1H
1I
1J
2
2A
2B
2C
2D
2E
2F
2G
2H
2I
2J
3
3A
3B
3C
3D
3E
3F
3G
5.1 WP 1: static work tasks
This WP runs over the complete 14-month duration of the project, including ten smaller WPs. The
WP starts with the design of the study design (A), followed by some pilot measurements (B). When
the final study design is set, subjects will be recruited (C) who will perform the experiment (D).
During data collection, a preliminary analysis (E) will be performed to get insights into data analyses
and provide the project partners with an interim report. At the end of data collection, the main
Research Proposal ADVANCE Page 9
analyses (F) will be performed, which are needed to evaluate the data and prepare them for
publication (G). At the beginning of the project, the experimenters should make sure that all
equipment is available or bought (H) and that there is a procedure to pay the participants (I). Along
the project, the complete project team will meet a couple of times (J) and some money is reserved to
visit national and international conferences (J).
5.2 WP 2: dynamic work tasks
This WP runs exactly parallel to WP 1, because the experiment will be designed so that all
participants perform a combination of one static and one dynamic simulated work element. This
means that the same subset of smaller WPs accounts in the same way as for WP.
5.3 WP 3: lumbar spine model
This WP is performed not by the Institute of Occupational and Social Medicine and Health Services
Research, but by the research cluster Simulation Technology of the group Biomechanics and
Biorobotics as part of the University of Stuttgart. As in the other two WP’s, this WP also relies on
measurement data, meaning that input for the modelling is considered both in the development of
the study design (A) and during the pilot measurements (B). The model of the lumbar spine already is
available, but needs to be adjusted so that information of the Laevo® exoskeleton can be
incorporated (C). When data collection has started, the collected data can be used to improve and
validate the model (D). As for WP 1 and WP 2, the appropriate equipment has to be available or
bought (E), participants should receive a refund (F), and knowledge gained during this project will be
shared with the project team (G) and with colleague scientists at national and international
conferences (G).
Research Proposal ADVANCE Page 10
6 Study procedure
6.1 Experimental tasks
In preparation to the measurements, we determined the details of the static and dynamic tasks in
combination with the trunk angles. The static task will mimic bimanual sorting in a 40° forward bent
position or a 40° forward bent position with 45° left or right rotation of the upper body (Figure 2).
The working distance and working height will be calculated by the individual’s body segment lengths,
the given forward trunk flexion angle of 40°, the resulting shoulder angle of 40° (holding the upper
arm orthogonal to the bas), and a defined elbow angle of 135°, counted from the base below the
ankle position. The working distance to the side is ~20 cm also counted from the position of the
ankle. The static tasks will be performed for 1.5 min in each position, with a passive break of 30 s in
between successive positions (total duration is 4.5 min, Figure 2B). In between consecutive blocks
(i.e., exoskeleton status and knee angle), participants get a passive break of ~2 min. This procedure is
repeated two times for exoskeleton status (with vs. without), lasting ~11 min in total.
Figure 2. (A) The static task performed in three positions: trunk forward flexion (left), trunk forward flexion with rotation to
the left (middle), and trunk forward flexion with rotation to the right (right). (B) The timeline for each exoskeleton status
(with vs. without), within which three working postures are measured, separated by 30-s passive breaks.
The dynamic task will mimic a bimanual box replacing task in a 70° forward bent position or a 70°
forward bent position with 45° rotation of the upper body (Figure 3A). The box handled will be ~10
kg with grips on both short sides, holding an arm angle of 135°. The working distance to the front is
~50 cm and to the side ~50 cm counted from the position of the ankle. The dynamic task will be
performed as follows: the box has to be taken from the height of the bending position (70°), moved
to an upright standing position, and placed back in bending position (70°) without or with rotation
sideward. The dynamic task will be performed in two series of 5 repetitions in each working posture
(forward trunk flexion, trunk rotation to the left, trunk rotation to the right), lasting 25 s per
repetition, with a 35-s passive break s after each series of 5 repetitions and a 60-s passive break after
each working position (total duration is 6.5 min). In between consecutive blocks (i.e., exoskeleton
status and knee angle), participants get a passive break of ~3 min. This complete procedure (Figure
3B) is repeated four times for exoskeleton status (with vs. without) and knee angle (flexion vs.
extension), all in all lasting ~35 min.
Research Proposal ADVANCE Page 11
Figure 3. (A) The dynamic task performed in three positions: trunk forward flexion (left), trunk forward flexion with rotation
to the left (middle), and trunk forward flexion with rotation to the right (right). (B) The timeline for a combination of
exoskeleton status (with vs. without) and knee angle (extended vs. flexed), within which three weights are handled,
separated by 60-s passive breaks.
6.2 Modelling of the lumbar spine
A lumbar spine model has been developed by the research group “Biomechanics and Biorobotics” of
the research cluster “Simulation Technology (SimTech)” of the University of Stuttgart, which can
predict internal forces during flexion movements (Rupp et al., 2015). The structure and properties of
the model are outlined in Figure 4, which includes a detailed lumbar spine with non-linear discs,
ligaments, and muscles.
Figure 4. Model of the lumbar spine (Rupp et al., 2015).
With the lumbar spine model, we aim to investigate how internal forces in the lumbar spine change
as a result of external forces of an exoskeleton during trunk flexion. This is the first time that an
Research Proposal ADVANCE Page 12
exoskeleton is coupled to a detailed biomechanical model of the lumbar spine. We will also try to
elaborate the modelling of the static task elements in trunk flexion and rotation by additional
analyses of dynamic trunk flexion movements.
6.3 Static trunk forward flexion angle while holding a load
From a technical point of view, we are interested in the natural relationship (i.e. without Laevo®)
between the spinal angle and the muscular stress on the back muscle (e.g., M. erector spinae
longissimus lumbalis). In addition, we are interested how this relationship changes when subjects are
wearing the exoskeleton, and when carrying different loads. Participants will perform a set of twelve
different tests, which are a combination of exoskeleton status (two levels: without vs. with Laevo®),
knee angle (extension vs. flexion), static trunk forward flexion angle (five levels; 0°, 30°, 60°, 60°, 30°),
and load carried (three levels; 0kg, 8kg, 16kg). The trunk flexion angles will be performed in
succession, meaning that a person will start upright (0°), slowly and uniformly move towards a slight
forward flexion (30°), slowly and uniformly move in a deeper forward flexion (60°), go deeper and
slowly and uniformly move back to the deep forward flexion (60°), and slowly and uniformly move
back into a slight forward flexion (30°) (Figure 5A). Using a full-factorial design with the independent
variables exoskeleton status and knee angle, participants will perform a total of four blocks. Within
each block, the load carried is performed for the trunk flexion angles, lasting ~30 s. In between
consecutive conditions within a block (i.e., carried weight), participants get a passive break of ~30 s
(total duration is ~2.5 min). In between consecutive blocks (i.e., exoskeleton status and knee angle),
participants get a passive break of ~3 min. This procedure is repeated four times for exoskeleton
status (with vs. without) and knee angle (extended vs. flexed), all in all lasting ~20 min (Figure 5B).
Figure 5. (A) Different trunk forward flexion angles while holding different loads, performed with and without the Laevo®
exoskeleton, with extended and flexed knees, while holding loads of 0kg, 8kg or 16kg. (B) The timeline for a combination of
exoskeleton status (with vs. without) and knee angle (extended vs. flexed), within which three trunk flexion angles are
measured in ascending and descending order, separated by 30-s passive breaks.
The arms will always be extended, and the weight will be bimanually carried, meaning that the total
weight (8kg and 16kg) will dividing over both hands (2×4kg and 2×8kg). The participant will receive
direct visual feedback on a monitor of the trunk flexion angle.
Research Proposal ADVANCE Page 13
6.4 Functional tasks
Functional tests are usually used for patients and elderly to estimate their level of performance. We
use a small selection of available functional tests to see how participants perform with and without
wearing the Laevo® exoskeleton. The following three tests are selected, each of which has its own
outcome with interpretation:
1. Stair Climb Test (SCT): ascend and descend an isolated set of 5 or 9 stairs (18 cm) using a
single handrail is measured (Bennell et al., 2011));
a. Outcome: time (seconds) taken to complete the number of steps negotiated;
b. Interpretation: smaller values represent better performance, and number of steps
negotiated in set time, where larger values represent better performance.
2. Timed Up & Go (TUG): assessment of a basic mobility skill (Bennell et al., 2011). The subject
is sitting on a regular chair with arm rests, rises from the chair, walks 3 m to the front, turns,
walks back to the chair, and sits down. The subject is wearing regular footwear (Bennell et
al., 2011; Dobson et al., 2013);
a. Outcome: time (s) taken to rise from a chair, walk 3 m, turn, walk back to the chair,
then sit down;
b. Interpretation: smaller values (faster time) represent better performance.
3. Course: containing three different work station simulations; (A) a static screwing task in a 40°
trunk flexion angle; (B) a dynamic logistic pick-and-place task replacing boxes from one side
to another; (C) a dynamic lifting task moving objects from a grid box to a table.
a. Outcome: time (s) taken to complete the course three times after another without
any breaks and heart rate (Hz) will be continuously recorded;
b. Interpretation: smaller time values (faster time) represent better performance;
smaller heart rate values represent less strain.
The functional tests are performed in a predefined order: (1) SCT, (2) TUG, (3) course. The order of
exoskeleton status will be randomized by drawing lots. For the SCT and TUG tests, the exoskeleton
will be worn but turned off; for the course the exoskeleton will be worn and turned on.
6.5 Experimental procedure
Participants will visit the work physiology laboratory at the Institute of Occupational and Social
Medicine and Health Services Research two times. During the first visit, the participants will be
officially included in the study after they have read the general information (Attachments 3), have
signed the informed consent form (Attachment 4), and have met the eligibility criteria (Attachment
5). When the participant is included, he will be familiarized with the Laevo® exoskeleton, the
simulated static and dynamic assembly tasks, and the functional tests. Finally, the participant will fill
out two short questionnaires providing some general information (Attachment 6) and information
about his history of musculoskeletal complaints (Attachment 7). The first visit will take ~1 hour.
During the second visit, the participants will perform the experiment, taking ~4 hours in total. The
study procedure is displayed in Figure 6. First, participants will be prepared with the measurement
equipment, i.e. electrodes will be placed to record muscle activity and heart rate and position
sensors to record postures and movements. After preparation, the participants have to perform
reference measurements to normalize the recordings of muscle activity (electromyography; EMG)
Research Proposal ADVANCE Page 14
and to correct the possible systematic error in the recordings of the body posture. The experiment
consists of three phases: first, the participants will perform the six experimental conditions (with vs.
without Laevo®, static vs. dynamic, knees extended vs. knees flexed); second, the participants will
perform the twelve angle-force conditions, grouped into four blocks (with vs. without Laevo®, knees
extended vs. flexed); third and last, the participant will perform the three functional tests. In
between each of these three sets of measurements, the participant will have a passive break of 5 min
and is allowed to drink or eat. After the static and dynamic tasks, participants are asked to fill out a
questionnaire (combination of NASA-TLX and self-developed questions; see Chapter 7.9). After all the
measurements are finished, the participants are asked to fill out a final questionnaire evaluating the
usability of the Laevo® exoskeleton (combination of SUS, TUI and self-developed questions; see
Chapter 7.9).
Figure 6. The experimental protocol is presented along with a timeline (min). Note that the main experimental conditions
are displayed in red boxes and the functional tests in blue boxes. The order of both sets is randomized across participants as
described previously. The test to determine the relation between postural angle and muscular stress is displayed in a black
box, which will also be performed in a randomized order. The yellow boxes represent the start and end of the complete
experiment, i.e. preparing and tidying up the participants.
6.6 Output
The collected data will be analyzed and prepared for statistical analyses if possible. All results will be
presented to the cooperation partners in workshops and/or interim and final reports. In addition, the
results will also be presented at symposia (e.g., annual GfA-congress, annual DGAUM- congress,
annual FAP- congress) and published in peer-reviewed international journals (e.g., triennial PREMUS-
congress, triennial IEA-congress, triennial ICOH-congress), to make the results accessible to the
scientific community.
Research Proposal ADVANCE Page 15
7 Measurement methods and analyses
7.1 General data
At the first visit to the laboratory, participants have to fill out the Standardized Nordic Questionnaire
(Nordic Musculoskeletal Questionnaire) to provide us with information about their musculoskeletal
history and some basic anthropometrics, including age, sex, body height, and body weight
(Attachments 6 & 7).
7.2 Muscle activity
The Laevo® exoskeleton aims to reduce the physical load on the back of the wearer, but it might
influence the physical load to other body parts such as the shoulders and legs. To record the activity
of selected muscles at all these body parts (Table 2), we will use surface EMG. This method uses
recording electrodes (42�24 mm, KendallTM H93SG ECG Electrodes, Covidien, Zaltbommel, the
Netherlands) that are attached to the skin’s surface overlaying the muscle of interest with an inter
electrode distance (IED) of 26 mm. The differential signal (bipolar derivation) of two electrodes
placed over one muscle belly recorded by an amplifier (PS12, THUMEDI® GmbH & Co. KG, Thum-
Jahnsbach, Germany), gives a firm estimation of that muscle’s electrical activity in microvolts (µV).
This electrical activity pattern enables us to estimate the magnitude of the muscle activity
(amplitude) and the overall firing frequency of the muscle (power frequency). The amplitude can be
calculated by the overall root-mean-square (RMS) of the power spectrum of the EMG signal. The
median power frequency (MPF) can be calculated as the median of the power spectral density of the
EMG signal.
7.3 Kinematics: posture and force
Combining ground reaction forces with linear segment accelerations and joint angles (and their
derived angular acceleration) enables the estimation of compression forces and force moments in
joints. Our joint of interest is the knee, of which we will estimate the compression forces and force
moments during the forward bent situations; the situations in which there is strong trunk rotation,
estimations may become inaccurate and unreliable. Several approaches exist to estimate joint forces
and joint moments, including forward dynamics, static body analyses, and inverse dynamics. We will
use inverse dynamics to estimate the forces and moments around the knee, for which we need the
(1) position of the lower and upper legs, (2) the ground reaction forces below the feet, (3) the
angular accelerations of the ankle and knee joints, and (4) segment parameters like segment length,
segment weight, and segment moment of inertia.
For the position, we will use position sensors that are integrated in modules of the PS12 system
(Thumedi® GmbH & Co. KG, Thum-Jahnsbach, Germany). The position sensors will be placed on the
lower leg, upper leg, and back providing us with the position in flexion and lateral flexion angles each
of the three segments.
Ground reaction forces below the feet will be recorded using a force plate (Bertec 3D-
Kraftmessplatte FP9090-15-1000, Velamed GmbH, Köln, Germany) with digital amplifier (Bertec,
AM6500, Velamed GmbH, Köln, Germany). The segment parameters will partly be measured
(segment lengths) and partly be estimated (segment weights and moments of inertia) using the four
summary tables as provided by Krishnan et al. (2016).
Research Proposal ADVANCE Page 16
Table 2. Six muscles of interest with a description of the posture and force level of their reference voluntary contraction
(RVC) and maximal voluntary contraction (MVC). The RVC will be performed once for ~10 s, with 1-min rest intervals in
between contractions of the same muscle and no rest in between RVCs and MVCs of different muscles.
Segment Muscle Reference voluntary contraction Force level
Lower
leg
M. gastrocnemius
medialis
Stand upright, unilateral isometric maximum planter flexion
contraction, i.e. going to the maximum / tip of the toes with full
strength (MVC) (Rieman et al., 2011)
Weight
own body
Upper
leg
M. vastus lateralis Lay prone, 90° knee flexion, foot neutral (90° ankle flexion), keep
position while a rope with a weight hanging over a pulley was attached
around the front-side of the ankle
5 kg
10 kg
M. biceps femoris Lay prone, 90° knee flexion, foot neutral (90° ankle flexion), keep
position while a rope with a weight hanging over a pulley was attached
around the back-side of the ankle
5 kg
7 kg
Lower
back
M. erector spinae
longissimus lumbalis
Lay prone, upper body and hips off the bench and extended legs fixed
with straps, arms hang orthogonal to the ground below the shoulder.
(1) Back (hip) extension without resistance (Abboud et al., 2014)
(2) Back (hip) extension with 15-kg resistance
(3) Back (hip) extension against fixed barrier (MVC)
Weight own
upper body
(+ 15 kg)
Abdomen
M. rectus abdominis Lay supine, upper body and hips off the bench and extended legs fixed
with straps, arms cross in front of the chest.
(1) Back (hip) flexion without resistance
(2) Back (hip) flexion with 10-kg resistance
Weight own
upper body
(+ 10 kg)
Shoulder/
neck
M. trapezius
descendens
Stand upright; arms in 90° arm abduction and slightly in the frontal
plane, elbows extended (not overstretched), holding a weight in each
hand
2×0 kg
2×2 kg
The angular accelerations can be derived from joint angles, which we will record using position
sensors that are also integrated in modules of the PS12 system (Thumedi® GmbH & Co. KG, Thum-
Jahnsbach, Germany). The position sensors will be placed on the foot, lower leg, and upper leg,
providing us with the flexion and lateral flexion angles of each segment with respect to the earth’s
radius of gyration. For joint angles, we can subtract one segment angle from another. In addition, we
will also record position of the spine by placing one sensor at the fifth lumbar vertebra (L5) and one
at the third thoracic vertebra (T3).
For the inverse modeling calculations, we will use a 2D model in which joints are represented as
hinges and segments as rigid bodies (Figure 7). We are interested in the net moment of the upper leg
around the lower leg (��, in Figure 7) and the force of the upper leg acting on the lower leg (��⃗ �, in
Figure 7).
The contact pressure between the Laevo® exoskeleton as exerted on the chest will be measured
using a Ø38 mm × 10 mm force sensor (Type KM38, ME-Meßsysteme GmbH, Henningsdorf,
Germany) that will be integrated in the chest pad.
Research Proposal ADVANCE Page 17
Figure 7. Illustrative inverse dynamic segment model of the lower leg and foot (left) with a sketch of the free body diagrams
of the foot (middle) and lower leg (right). The symbols represent kinematic parameters: ankle (�), knee ( ), center of mass
of the foot (�), center of mass of the lower leg (�), flexion angle of the lower leg (�), ground reaction force (��⃗ �,), force
of the lower leg acting on the foot (��⃗ ,), force of the foot acting on the lower leg (��⃗ ,), force of the upper leg acting on
the lower leg (��⃗ �,), gravitational force of the foot (���⃗ ), gravitational force of the lower leg (���⃗ ), net moment of lower leg
around foot (�,), net moment of foot around lower leg (�,), and net moment of upper leg around lower leg (��,).
7.4 Reference measurements
A reference posture will be recorded while the subject is standing upright with his back straight
against a vertically levelled wall, his arms alongside his body, and his hand palms facing forward. The
angles from this recording all represent zero angles, and will be used to correct the experimental
recordings for any systematic offsets.
EMG is recorded in microvolt (µV) and may vary considerably between subjects. For this reason, we
will normalize the activity of all measured muscles to a reference voluntary electrical activity (RVE)
assessed during reference voluntary contraction (RVC; Table 2). During the reference contraction,
the experimenter encourages the subject to stay in the reference posture. Each reference
contraction will take ~10 s and will be performed once, of which the most stable 5-s plateau of one of
the contractions will be used for EMG normalization and referred to as percent reference voluntary
electrical activity (%RVE; Mathiassen et al., 1995). The reference measurements take ~30 min in
total.
7.5 Heart rate
The heart rate will be continuously recorded by electrocardiography (ECG) using ECG recording
electrodes (42�24 mm, KendallTM H93SG ECG Electrodes, Covidien, Zaltbommel, the Netherlands).
The two electrodes will be attached to the skin’s surface, placed ~5 cm cranial and ~3 cm left-lateral
from the distal end of the sternum and over the anterior to midaxillary line at the fifth left rib. The
heart rate signal will be recorded by the same amplifier as EMG (PS12-UD, THUMEDI® GmbH & Co.
KG, Thum, Germany).
7.6 Discomfort
Participants will rate their overall subjective feeling of discomfort (RPD) on an 11-point numerical
rating scale (McCaffery and Beebe, 1989) (Attachments 8 & 9), ranging from 0 (no discomfort at all)
to 10 (maximally imaginable discomfort). When participants experience a feeling of discomfort > 0,
we ask them to define the body region (Kuorinka et al., 1987) causing this sensation.
Research Proposal ADVANCE Page 18
7.7 Functional tests
To assess how wearers of the Laevo® can walk around and use the exoskeleton in between work
tasks, a set of two functional tests will be assessed by measuring time duration and/or difficulty
and/or on an 11-point numeric rating scale. For each functional test, the outcome(s) to be recorded
is listed below:
1. Stair Climb Test (SCT): time (s) taken to complete the number of steps negotiated (Bennell et
al., 2011);
2. Timed Up & Go (TUG): time (s) needed to rise up from the chair, walk, turn, and sit back
down on the chair again (Dobson et al., 2013).
3. Course: time (s) needed to complete the course and average heart rate (Hz) over the course.
7.8 Relation static trunk forward flexion angle, carried load and
muscular stress / spinal load
With the data collected from the twelve tests that last ~30 s, we will be able to draw curves for every
load carried across different trunk forward flexion angles. These curves will provide us with
information whether the relation between increased trunk forward flexion is linear or not with
respect to outcomes such as muscular activity of the lower back muscle (i.e., M. erector spinae
longissimus lumbalis) or the abdomen muscle (i.e., M. rectus abdominis). In addition, separate lines
within these graphs account for exoskeleton status (with vs. without exoskeleton), knee angle
(extension vs. flexion) and weight carried (0kg vs. 8kg vs. 16kg), during which the relation between
the trunk flexion angle and the muscular activity level could change too.
7.9 Participant evaluation
To evaluate the workload of the complete set of experimental conditions, we will use the NASA-TLX
(Hart and Staveland, 1988; Attachment 10), which assesses six dimensions of workload (i.e., mental
demand, physical demand, temporal demand, own performance, effort, and frustration). Subjects
will rate only three of these dimensions on a scale from 0 to 100 (physical demand, temporal
demand, and effort). We will assess overall partial workload by averaging or adding the scores of the
three dimensions (Hoonakker et al., 2011). In addition, we will analyze the scores of the subscales
(Hart, 2006; Hoonakker et al., 2011). Next to the NASA-TLX, we will also ask eight self-developed
questions (Attachment 10).
To evaluate participants view on the usability and acceptance of the Laevo® exoskeleton in the
performed tasks and on the acceptance of the Laevo® exoskeleton in work settings, we will ask six
self-developed questions, use the ten questions of the standardized and validated questionnaire
System Usability Scale (SUS), and use a selection of seven questions from the standardized
Technology Usage Inventory (TUI) from the domains usability (n = 3) and skepticism (n = 4). The final
set of questions (n = 23; see Attachment 11) will take approximately 10 minutes.
The SUS is a quick and easy to use method to collect the usability evaluation of users of a system
(Borsci et al., 2009), in this case the system is the Laevo® exoskeleton. The SUS is also seen as a tool
to measure the subjective perception of interaction with a system (Brooke, 1996). The SUS is
recommended as satisfaction metric to complement other performance metrics such as time on task,
Research Proposal ADVANCE Page 19
which in this study is done with the functional tests. The assessment of the SUS is the level of
agreement on a Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) (Brooke, 1996).
The TUI is a questionnaire that can be used for research purposes or in the field of technology
acceptance research (Kothgassner et al., 2013). It could support technique developers in the user-
centered evaluation of new technologies, such as virtual simulations. The TUI records technology-
specific and psychological factors that contribute to the actual use of the technology, containing the
following eight scales: curiosity, anxiety, interest, usability, immersion, usefulness, skepticism, and
accessibility. The TUI also includes the Intention to Use (ITU) scale, which contains three questions.
The assessment of the TUI is the level of agreement on a Likert scale ranging from 1 (not true) to 7
(true). From the original TUI, we will only include the questions from the domains usability (n = 3)
and skepticism (n = 4).
7.11 Statistical analysis
This study aims to answer the research question “what is the influence of exoskeleton, task, and
posture on muscular activity, knee joint loading, spinal loading, and discomfort?”. Using a within-
subject design, i.e. a randomized, balanced cross-over design, the data should be analyzed using a
repeated-measures analysis of variance (ANOVA). This model contains three independent, within-
subject factors: exoskeleton (with vs. without), posture (forward bend vs. rotation), and task (static
vs. dynamic). The outcome parameters include muscular activity (i.e. RMS of EMG) of the selected
muscles (Table 2), knee compression force, spinal loading, and discomfort. We will use JMP software
(JMP® 13.1.0, SAS Inc., Cary, NC, USA) or SPSS software (IBM® Statistics® 25.0, Armonk, NY, USA) to
perform the repeated measures ANOVA and accept statistical significance for p < 0.05.
The outcomes of the functional tests are based on numeric rating scales and / or are time-related,
and will be statistically evaluated using a paired T-Test to compare the conditions with and without
Laevo®. The results from the tests combining exoskeleton with trunk forward flexion angle and with
load carried will be plotted to see if the relationship between the trunk forward flexion angle (x-axis)
and the muscular stress of the erector spinae muscle (y-axis) changes per load carried (different
lines) and exoskeleton (with vs. without; different lines).
The relationships between static trunk forward flexion angle and muscular activity will be assessed
using regression analysis and compared between different loads carried (while keeping the
exoskeleton status constant) and between exoskeleton statuses (while keeping the carried load
constant).
The results of the two questionnaires that evaluate the acceptability and usability of the Laevo®
exoskeleton will be narratively discussed.
Research Proposal ADVANCE Page 20
8 Risk for the participant
The experiment has no threatening risks for the participant and all measurement equipment is
certified in declarations of conformity (Attachments 12-16). Note that there can be some side-effects
of study-related procedures about which we will inform the participant, including:
- Skin irritation or reddening due to skin preparation and of the attachment of the adhesive
electrodes; in rare cases, participants may experience allergic reactions;
- When the electrode cables are tights and pull the skin, participants could experience tingling
or tension sensations in the area of the adhesive electrodes due to this cable pull;
- Sensations of tension or muscle fatigue along the duration of the experiment;
- Due to participants performing the experiment have to stand during the measurements, it
might occur that participants experience possible sensations of dizziness, loss of blood
pressure, and a very low risk of impotence due to the ~60-min periods of interrupted
standing;
- Minor risk of muscle aching the day after the experiments due to maximum voluntary muscle
contraction.
Research Proposal ADVANCE Page 21
9 Data protection
The participants as well as their personal and recorded data will be numerically pseudonymized by
assigning a randomly generated two-digit identification number to the examined participant (e.g.,
S30P_ _). Relating analyzed outcome data to the natural person is only possible by inspection of the
decoding list. This list, together with the informed consents of all volunteering participants, will be
stored separately from the measured data in the storage of the Institute of Occupational and Social
Medicine and Health Services Research. All participant data will be stored for a period of ten years
after publication of the results. The destruction of the data in paper form is carried out by means of
the disposal boxes for data protection at the University Hospital, and the destruction of the data in
electronic form is carried out by means of transferring the data off the memory cards of the devices.
The measurements as part of this proposal are all recorded on portable computers that are
accessible by means of a password only and cannot be accessed by any network. The pseudonymized
data are processed within the University Hospital (UKT) network, of which the time of the
measurement collection is removed from the operating software whenever possible. The measuring
devices are configured identically for all measurements to exclude a possible association between
the measurement configuration and the participants. After completing the data analysis, the data of
the participants are stored together with the photo documentation on Blu-ray discs.
Only employees directly involved in the study have access to the data. In addition, all employees of
the Institute for Occupational and Social Medicine and Health Services Research who are involved in
the study are subject to medical confidentiality.
At the start of the study, participants are asked whether they wish to be informed about the primary
results of the study. In case of several interested parties, an information event will be organized for
the participants after completion of the data evaluation. If there is only little interest, a brief written
presentation of the primary results will be spread and sent to interested parties by e-mail.
Research Proposal ADVANCE Page 22
10 Study finance
The financing of the study, including personnel costs and facility costs for technical equipment and
assistance, are funded by several industrial companies (Table 3). The study participants will receive a
compensation of about € 10 per hour. The familiarization day is without refund, the testing day will
take approximately 4 to 5 hours, which results in a total amount of € 45.
Table 3. Funding of the project is divided over eight partners, although not every partner collaborates in every work
package.
Work package
Partner
WP 1
Static task element
WP 2
Dynamic task element
WP 3
Lumbar spine modelling
Daimler AG
AUDI AG
BMW AG
MTU Aero Engines AG
Iturri Gruppe
Deutsche Post AG
BASF
Dachser Intelligent Logistics
Research Proposal ADVANCE Page 23
11 Personnel Staff
In work packages (WP) 1 and 2, employees from the Institute of Occupational and Social Medicine
and Health Services Research of the University Hospital Tübingen & University of Tübingen will be
responsible for each of the steps. In WP 3, all the modelling work is done by the group Biomechanics
and Biorobotics of the cluster Simulation Technology (SimTech) of the University of Stuttgart.
11.1 Work packages 1 & 2
The main contributors to these two work packages are:
- Benjamin Steinhilber (15% of 100%-employment = 15%)
o Project leader: responsible for the study design, data interpretation;
- Tessy Luger (15% of 100%-employment = 15%)
o Project leader: responsible for the study design and supervising the following stages
of the study;
- Robert Seibt (100% of 65%-employment = 65%)
o Technician and researcher: responsible for study design, measurement equipment,
pilot and final measurements, data analysis, and data interpretation;
- Marlene Röttgen (6.7% of 75%-employment = 5%)
o Project management;
- PhD student, Mona Bär (100% of 65%-employment = 65%)
o Preparing, performing, analyzing and interpreting measurements.
11.2 Work package 3
- Benjamin Steinhilber (15% of 100%-employment = 15%)
o Project leader: responsible for the study design, data interpretation;
- Tessy Luger (15% of 100%-employment = 15%)
o Project leader: responsible for the study design and supervising the following stages
of the study;
- Marlene Röttgen (6.7% of 75%-employment = 5%)
o Project management;
- PhD student, Mona Bär (100% of 65%-employment = 65%)
o Model building and validation.
Research Proposal ADVANCE Page 24
11 References
Abboud, J., Nougarou, F., Page, I., Cantin, V., Massicotte, D., Descarreaux, M., 2014.
Trunk motor variability in patients
with non-specific chronic low back
pain. European Journal of Applied
Physiology 114 (12), 2645-2654. doi:
10.1007/s00421-014-2985-8
Allison, G.T., Godfrey, P., Robinson, G., 1998.
EMG signal amplitude assessment
during abdominal bracing and
hollowing. Journal of Electromyography and Kinesiology 8
(1), 51-57.
Bao, S., Mathiassen, S.E., Winkel, J., 1995.
Normalizing upper trapezius EMG
amplitude: comparison of different
procedures. Journal of
Electromyography and Kinesiology 5
(4), 251-257.
Bate, S.T., Jones, B., 2006. The Construction of
Nearly Balanced and Nearly Strongly Balanced Uniform Cross-Over Designs.
Journal of Statistical Planning and
Inference 136, 3248-3267.
Bennell, K., Dobson, F., Hinman, R., 2011.
Measures of physical performance
assessments: Self-Paced Walk Test
(SPWT), Stair Climb Test (SCT), Six-
Minute Walk Test (6MWT), Chair
Stand Test (CST), Timed Up & Go
(TUG), Sock Test, Lift and Carry Test (LCT), and Car Task. Arthritis Care Res
(Hoboken) 63 Suppl 11, S350-370. doi:
10.1002/acr.20538
Borsci, S., Federici, S., Lauriola, M., 2009. On
the dimensionality of the System
Usability Scale: a test of alternative
measurement models. Cogn Process
10 (3), 193-197. doi: 10.1007/s10339-
009-0268-9 Bosch, T., van Eck, J., Knitel, K., de Looze, M.,
2016. The effects of a passive
exoskeleton on muscle activity,
discomfort and endurance time in
forward bending work. Applied
Ergonomics 54, 212-217. doi:
10.1016/j.apergo.2015.12.003
Brooke, J., 1996. SUS - A quick and dirty
usability scale, in: Jordan, P.W.,
Thomas, B., Weerdmeester, B.A., McClelland, A.L. (Eds.), Usability
evaluation in industry. Taylor and
Francis, London, UK.
Cordasco, F.A., Wolfe, I.N., Wootten, M.E.,
Bigliani, L.U., 1996. An
electromyographic analysis of the
shoulder during a medicine ball
rehabilitation program. The American
Journal of Sports Medicine 24 (3), 386-
392. doi: 10.1177/036354659602400323
Cormie, P., McGuigan, M.R., Newton, R.U.,
2010. Changes in the eccentric phase
contribute to improved stretch-
shorten cycle performance after
training. Medicine and Science in
Sports Exercise 42 (9), 1731-1744. doi:
10.1249/MSS.0b013e3181d392e8
de Looze, M.P., Bosch, T., Krause, F., Stadler,
K.S., O'Sullivan, L.W., 2015. Exoskeletons for industrial application
and their potential effects on physical
work load. Ergonomics 59 (5), 671-
681. doi:
10.1080/00140139.2015.1081988
Dobson, F., Bennell, K.L., Hinman, R.S., Abbott,
J.H., Roos, E.M., 2013. Recommended
performance-based tests to assess
physical function in people diagnosed
with hip or knee osteoarthritis. University of Melbourne, Melbourne,
Australia.
Hart, S.G., 2006. NASA-Task Load Index (NASA-
TLX); 20 years later, Proceedings of
the Human Factors and Ergonomics
Society 50th Annual Meeting. Human
Factors and Ergonomics Society, San
Fransisco, California, US, pp. 904-908.
Hart, S.G., Staveland, L.E., 1988. Development of NASA-TLX (Task Load Index):
Results of Empirical and Theoretical
Research. Advances in Psychology 52,
139-183. doi: 10.1016/S0166-
4115(08)62386-9
Hoonakker, P., Carayon, P., Gurses, A., Brown,
R., McGuire, K., Khunlertkit, A.,
Walker, J.M., 2011. Measuring
Workload of Icu Nurses with a
Research Proposal ADVANCE Page 25
Questionnaire Survey: The Nasa Task
Load Index (Tlx). IIE Transactions on
Healthcare Systems Engineering 1 (2),
131-143. doi:
10.1080/19488300.2011.609524 Irwin, R.W., Zuhosky, J.P., Sullivan, W.J., Foye,
P.M., Sable, A.W., Panagos, A., 2007.
Industrial Medicine and Acute
Musculoskeletal Rehabilitation. 5.
Interventional Procedures for Work-
Related Lumbar Spine Conditions.
Archives of Physical Medicine and
Rehabilitation 88 (3 (Suppl. 1)), S22-
S28. doi: 10.1016/j.apmr.2006.12.012
Ito, K., Nonaka, K., Ogaya, S., Ogi, A., Matsunaka, C., Horie, J., 2016. Surface
electromyography activity of the
rectus abdominis, internal oblique,
and external oblique muscles during
forced expiration in healthy adults.
Journal of Electromyography and
Kinesiology 28, 76-81. doi:
10.1016/j.jelekin.2016.03.007
Juker, D., McGill, S., Kropf, P., Steffen, T.,
1998. Quantitative intramuscular myoelectric activity of lumbar
portions of psoas and the abdominal
wall during a wide variety of tasks.
Medicine and Science in Sports
Exercise 30 (2), 301-310.
Kothgassner, O.D., Felnhofer, A., Hauk, N.,
Kastenhofer, E., Gomm, J., Kryspin-
Exner, I., 2013. Technology Usage
Inventory (TUI): Manual. ICARUS
Research Team, Wien. Krishnan, R.H., Devanandh, V., Brahma, A.K.,
Pugazhenthi, S., 2016. Estimation of
mass moment of inertia of human
body, when bending forward, for the
design of a self-transfer robotic
facility. Journal of Engineering Science
and Technology 11 (2), 166-176.
Kuorinka, I., Jonsson, B., Kilbom, A.,
Vinterberg, H., Biering-Sorensen, F.,
Andersson, G., Jorgensen, K., 1987. Standardised nordic questionnaires
for the analysis of musculoskeletal
symptoms. Applied Ergonomics 18 (3),
233-237.
Lee, H., Kim, W., Han, J., Han, C., 2012. The
technical trend of the exoskeleton
robot system for human power assistance. International Journal of
Precision Engineering and
Manufacturing 13 (8), 1491-1497. doi:
10.1007/s12541-012-0197-x
MacDougall, W., 2014. Industrie 4.0: Smart
Manufacturing for the Future. Asmuth
Druck & Crossmedia GmbH & Co. KG,
Cologne, Germany.
Mathiassen, S.E., Winkel, J., Hagg, G.M., 1995.
Normalization of Surface EMG Amplitude from the Upper Trapezius
Muscle in Ergonomic Studies - A
Review. Journal of Electromyography
and Kinesiology 5 (4), 197-226.
McCaffery, M., Beebe, A., 1989. Pain: Clinical
manual for nursing practice. Mosby St.
Louis, MO, USA.
Newton, R., 1989. Review of tests of standing
balance abilities. Brain Injury 3 (4),
335-343. doi: 10.3109/02699058909004558
Nyberg, L.A., Hellenius, M.L., Kowalski, J.,
Wandell, P., Andersson, P., Sundberg,
C.J., 2011. Repeatability and validity of
a standardised maximal step-up test
for leg function--a diagnostic accuracy
study. BMC Musculoskeletal Disorders
12, 191. doi: 10.1186/1471-2474-12-
191
Padulo, J., Tiloca, A., Powell, D., Granatelli, G., Bianco, A., Paoli, A., 2013. EMG
amplitude of the biceps femoris
during jumping compared to landing
movements. SpringerPlus 2, 520. doi:
10.1186/2193-1801-2-520
Reneman, M.F., Brouwer, S., Meinema, A.,
Dijkstra, P.U., Geertzen, J.H.B.,
Groothoff, J.W., 2004. Test–Retest
Reliability of the Isernhagen Work
Systems Functional Capacity Evaluation in Healthy Adults. Journal
of Occupational Rehabilitation, 14 (4),
295-305.
Research Proposal ADVANCE Page 26
Rieman, B.L., Limbaugh, G.K., Eitner, J.D.,
LeFavi, R.G., 2011. Medial and lateral
gastrocnemius activation differences
during heel-raise exercise with three
different foot positions. Journal of Strength and Conditioning Research
25 (3), 634-639. doi:
10.1519/JSC.0b013e3181cc22b8
Rupp, T.K., Ehlers, W., Karajan, N., Gunther,
M., Schmitt, S., 2015. A forward
dynamics simulation of human lumbar
spine flexion predicting the load
sharing of intervertebral discs,
ligaments, and muscles. Biomech
Model Mechanobiol 14 (5), 1081-1105. doi: 10.1007/s10237-015-0656-
2
Scott, N.A., Moga, C., Harstall, C., 2010.
Managing low back pain in the
primary care setting: The know-do
gap. Pain Research & Management 15
(6), 392-400.
Siddiqi, A., Arjunan, S.P., Kumar, D., 2015.
Improvement of isometric dorsiflexion
protocol for assessment of tibialis anterior muscle strength. MethodsX 2,
107-111. doi:
10.1016/j.mex.2015.02.006
Spada, S., Ghibaudo, L., Gilotta, S., Gastaldi, L.,
Cavatorta, M.P., 2017. Investigation
into the Applicability of a Passive
Upper-limb Exoskeleton in
Automotive Industry. Procedia
Manufacturing 11, 1255-1262. doi:
10.1016/j.promfg.2017.07.252 Tartaglia, G.M., Barozzi, S., Marin, F., Cesarani,
A., Ferrario, V.F., 2008.
Electromyographic activity of
sternocleidomastoid and masticatory
muscles in patients with vestibular
lesions. Journal of Applied Oral
Science 16 (6), 391-396.
van Rijn, R.M., Huisstede, B.M.A., Koes, B.W.,
Burdorf, A., 2010. Associations
between work-related factors and
specific disorders of the shoulder – a
systematic review of the literature.
Scandinavian Journal of Work,
Environment & Health 36 (3), 189-201.
doi: 10.5271/sjweh.2895 Vellas, B.J., Wayne, S.J., Romero, L.,
Baumgartner, R.N., Rubenstein, L.Z.,
Garry, P.J., 1997. One-leg balance is an
important predictor of injurious falls
in older persons. Journal of the
American Geriatrics Society 45 (6),
735-738.
Venkatesh, V., Davis, F.D., 2000. A Theoretical
Extension of the Technology
Acceptance Model: Four Longitudinal Field Studies. Management Science 46
(2), 186-204. doi:
10.1287/mnsc.46.2.186.11926
Viteckova, S., Kutilek, P., Jirina, M., 2013.
Wearable lower limb robotics: A
review. Biocybernetics and Biomedical
Engineering 33 (2), 96-105. doi:
10.1016/j.bbe.2013.03.005
Wall, R., Garcia, G., Läubli, T., Seibt, R., Rieger,
M.A., Martin, B., Steinhilber, B., In preparation. Physiological changes
during prolonged standing and
walking considering age, gender and
standing work experience.
Wang, X., Dong, X.S., Choi, S.D., Dement, J.,
2017. Work-related musculoskeletal
disorders among construction workers
in the United States from 1992 to
2014. Occup Environ Med 74 (5), 374-
380. doi: 10.1136/oemed-2016-103943
Yang, C.J., Zhang, J.F., Chen, Y., Dong, Y.M.,
Zhang, Y., 2008. A Review of
exoskeleton-type systems and their
key technologies. Proceedings of the
Institution of Mechanical Engineers,
Part C: Journal of Mechanical
Engineering Science 222 (8), 1599-
1612. doi: 10.1243/09544062jmes936